Intelligence is a rate, not a static quality. You can outperform someone who learns in fewer repetitions by simply executing your own (potentially more numerous) repetitions on a faster timeline. Compressing the time between attempts is a controllable way to become 'smarter' on a practical basis.
Related Insights
After initially modeling others, mastery comes from generating 'first-party data.' Execute a high volume of repetitions, then analyze your own top 10% of outcomes. Identify the observable differences between your best and worst results, incorporate those learnings, and repeat the cycle for a powerful, personalized feedback loop.
Even with vast training data, current AI models are far less sample-efficient than humans. This limits their ability to adapt and learn new skills on the fly. They resemble a perpetual new hire who can access information but lacks the deep, instinctual learning that comes from experience and weight updates.
The greatest performers, from athletes to companies, are not just the most talented; they are the best at getting better faster. An obsession with root-cause analysis and a non-defensive commitment to improvement is the key to reaching otherwise unachievable levels of success.
When hiring, prioritize a candidate's speed of learning over their initial experience. An inexperienced but rapidly improving employee will quickly surpass a more experienced but stagnant one. The key predictor of long-term value is not experience, but intelligence, defined as the rate of learning.
Many perceived failures, from business to dating, stem from a radical underestimation of the repetitions required for success. Most problems can be solved not by more talent, but by applying an unreasonable amount of volume.
While repetition is crucial for skill mastery, the brain eventually stops recording familiar experiences to conserve energy. This neurological efficiency causes our perception of time to speed up as we age. To counteract this, one must intentionally introduce new challenges to keep the brain actively creating new memories.
Don't get stuck trying to perfect your strategy. Commit to a high volume of action first. The pain of inefficiency from doing the work will naturally motivate you to learn and optimize your process, leading to mastery faster.
In rapidly evolving fields like AI, pre-existing experience can be a liability. The highest performers often possess high agency, energy, and learning speed, allowing them to adapt without needing to unlearn outdated habits.
For cutting-edge AI problems, innate curiosity and learning speed ("velocity") are more important than existing domain knowledge. Echoing Karpathy, a candidate with a track record of diving deep into complex topics, regardless of field, will outperform a skilled but less-driven specialist.
Simply practicing a new skill is inefficient. A more effective learning loop involves four steps: 1) Reflect to fully understand the concept, 2) Identify a meaningful application, 3) Practice in a low-stakes environment, and 4) Reflect again on what worked and what didn't to refine your approach.
